[Fwd: Evolution, and 'functional' + 'social']

Charles Li li at GRADDIV.UCSB.EDU
Wed Dec 4 18:03:14 UTC 2002


The issue of "boundary" between human (cognition, culture, language) and
animal (cognition, culture, communication), like any other issues
and  concepts in biology, is not subject to water-tight generalization.
This lack of theorem-like generalities in biological science marks its
fundamental difference from physics/mathematics. The difference has
profound implications on research methodologies and ways of thinking in
these two branches of science. Francis Crick has written an elegant book
(The Mad Pursuit of Knowledge)  on this issue based on his own experience
starting as a physicist and then becoming a biologist.

Talmy assumes a continuum between animal cognition, culture, communication
and human cognition, culture and language - a sweeping generalization
resplendent with Givonian speed and epistemology. For sure, there is a lot
of commonality as well as finely graded differences or continuum within a
Linnean order, say, mammals. For example, the body design, the
cytoarchitecture of the brain, the basic emotions (as opposed to cognitive
emotions). Within the confines of communication, facial expressions are
finely graded from lower mammals to higher mammals. The central issue here,
however, concerns language: Is there a clear and distinct boundary between
human language and animal communication? Conversely, is the difference
between human language and higher primate communication a finely graded
continuum like facial expressions?

There are several unbridgeable and distinct boundaries between animal
communicative behavior and the evolutionary development of hominid
communicative behavior leading toward language. The first one is the
emergence of symbolic signals. At the most elemental cognitive level, a
symbolic signal must possess two properties:
(a)     It refers to a concrete object
(b)     The reference is context-independent
This is the beginning of  'meaning' and vocabulary. The emergence of the
first symbolic signal referring to a concrete object represents the
crossing of the first 'missing link' in hominid evolution. Animal
communicative signals are not symbolic according to our definition of
elemental symbolic signals, and animal communicative signals do not have
'meaning'. What they have is 'function' such as threat, distress, begging,
appeasement/submission, courtship/copulation, warning/alarm, recruitment,
assembly, dispersion, identification, territoriality, feeding, etc.
Functions must not be confused with 'meaning".  Each of the following
linguistic expressions has the function of 'threat':
         I'm going to bite your ears.
         I'll kill you.
         I'll knock you over.
         I'm going to beat you up.
But they all have different meanings, and they have different meanings
because they are linguistic expressions. In animal communicative behavior,
threat signals are just threat signals, no more and no less. Different
species have different threat signals. Some signals are graded according to
the intensity of the signaler's emotional state, and some, discrete. Some
are structurally more complex involving various components and several
channels of communication, and some are simple. All members of the same
species use the same threat signals, and all threat signals are emitted by
signalers to threaten intended receivers. They have no meaning. We can make
similar statements about any other functional category of animal
communicative signals.
A particular set of animal communicative signals that have received a great
deal of attention is the frequently cited warning signals of vervet
monkeys. (Many other animals have different warning calls for different
approaching predators. Suricates, a burrowing viverrid related to the
mongoose is a good example.) These warning calls are not symbolic signals
because they do not possess property (b). Nevertheless, the warning calls
of the vervet monkeys represent a step toward a symbolic signal, because
they differentiate, for example, reptilian, avian, mammalian and other
predators. The differentiation, however, holds only in the context of
warning. It is not context-free. In other words, a vervet monkey warning
signal has no referential property outside of the context of warning.
Seyfarth & Cheney (1999) note that vocal production, i.e. delivery of
acoustically defined calls, among apes and monkeys appears fully formed
shortly after birth, suggesting that vocal production may be largely
innate. In addition, Aitken (1981) and Pandya et al (1988) conducted
experiments on capuchin monkeys showing that their vocal production was
mediated primarily by the central (periaqueductal) gray area of the
mid-brain, a phylogenetically very old set of neurons responsible for
arousal and motivational states in all vertebrates. In the case of vervet
monkey's warning calls, the only role of the neocortex involves associating
a particular involuntary vocalization with a specific situation. The
vocalization is involuntary because it is probably associated with fear
aroused by the situation. Hence an infant vervet possesses the adult
repertoire of vocalization. The learning during ontological development
involves the correct coupling of one involuntary vocalization with one
specific dangerous situation. Each vocalization can be graded according to
intensity. But it is only the coupling process that is mediated by the
neocortex, and this coupling process, according to Seyfarth & Cheney,
requires learning. The neural mechanism we have just sketched for non-human
primate communication contrasts sharply with the neural mechanism of the
production of causal, spoken language. The production of casual, spoken
language is primarily, though not exclusively, mediated by the neocortex.
The emotional/motivational state of the speaker can be viewed as a
coterminous but neurologically separate dimension of speech expressed
primarily in prosody. It is, therefore, not surprising that participants in
casual spoken conversation can talk about things that are remote in time
and space from the occurrence of the conversation. This is the
"displacement" feature of human language that Charles Hockett (1960)
pointed out. It does not exist in non-human primate communication because a
non-human primate communicative signal tends to be associated with the
emotional or motivational reaction to a particular situation including the
animal's own internal hormonal state. (The interesting exception here is
the waggle dance of homey bees. That is another story.)

Another set of animal communicative signals that might be construed as
symbolic are the signals that have the function of individual
identification. For example, the whistle of the bottlenose dolphin and the
vocalization of social mammals is acoustically unique for each individual.
Consequently members of a social group can identify an individual upon
detecting its vocalization. The identification function of such animal
communication signals parallels the identification function of the voice of
a human. Every human voice is acoustically unique. Within a social group,
members can identify each other by the voice of a person. The unique
quality of the voice, however, does not constitute a symbolic signal. It
does not have the referential property of the expression "It's me!" or the
referential property of a first person singular pronoun. Recognition of the
voice depends on past interaction and social familiarity. We cannot
identify a stranger's voice, although we recognize that it does not belong
to someone we know. The same can be said about the set of animal
communicative signals that have the function of identification. The whistle
of the bottle-nosed dolphin, the coo of a vervet monkey, the bark of a
wolf, the grunt of a warthog and the various other communicative signals
used by animals for maintaining contact within a social group are not
symbolic signals. They are neither words nor names.

In short, animal communicative signals in nature do not contain symbolic
signals. Without symbolic signals, there is no meaning. Meaning and
symbolic signals are at the heart of human language. Only symbolic signals
can be concatenated to form larger units of communicative signals, i.e.
morpho-syntactic and discourse units, and only symbolic signals can lead to
metaphors.

Talmy's point on the nature of language change is partly correct and partly
an attack of a straw-man by lambasting the "extreme functionalist position"
of attributing ALL language change to social and cultural factors. I don't
know of anyone holding such a point of view. My belief that language change
tends to be driven by social and cultural factors is based on the
hypothesis that the default situation of any language, prehistoric as well
as contemporary, is contact. Contact-induced changes are cultural and
socially driven. Unfortunately a great deal of contact-induced changes in a
specific language beyond a certain time depth cannot be reconstructed. One
piece of evidence in support of my hypothesis is areal features among
genetically unrelated languages. Now we are entering a different domain of
linguistics. Suffice to say that my main point is to clarify that language
change has nothing to do with life expectancy, genetic mutation and
reproductive success. The mainstay of evolution of animal communicative
signals is ritualization. Ritualization typically involves evolutionary
development making a signal more conspicuous, i.e. more strongly
stimulating, and ways of making a signal more stable, i.e. having a typical
form, and typically such evolutionary development occurs on an evolutionary
time scale, i.e. thousands of generations at a minimum, e.g. exaggeration
of movements, addition of brightly colored anatomical components, creation
of new exocrine glands, enlargement of anatomical components involved in
display, etc. In short, ritualization may involve anatomical, physiological
as well as behavioral changes that enhance the efficiency of the signal.
Hence, the addition of properties that are the consequence of direct
selection of increased effectiveness of the signal. Many of these
evolutionary changes of animal communicative signals involve genetic
mutations.
It is true one may couch human cultural development in Darwinian
evolutionary terms. Dawkins tried and proposed "emes" as the unit of
"selection" in cultural evolution, a homology of "genes" in biological
evolution, although genes are not the only entity on which natural
selection operates. Natural selection also operates on "organism" and
"species". But that is another topic. The relevant point here is that even
though one may wish to couch cultural development in Darwinian's
evolutionary theory, it does not follow that there isn't a quantum leap
from animal communicative behavior to human language.
Steve Long erroneously assumed that discontinuity between animal
communicative behavior and human language implied that human language
originated through mutation. Bickerton, Pinker, Richard Klien, and those
who subscribe to the hypothesis of hardwired universal grammar may believe
that human language originated through genetic mutation. I have argued that
the evolutionary process leading to the crystallization of language took
approximately one to one-and-half million years. There is no such thing as
a linguistic gene. A gene merely specifies the structure of a protein.
Language, however, is a behavior at the highest cognitive level. It
permeates all facets of human cognition. It is not and cannot be brought
about by a single or a few genes! Even a simple biological trait such as
the color of the eye requires a cascade of interactions among seven
different genes.  Any defective protein that severely impairs a person's
ability to acquire a language is bound to have dramatic consequences on the
ontological development of that person's central nervous system. This is
not to say that the evolutionary process leading to the crystallization of
language does not involve genetic mutations. On the contrary,  increased
encephalization, lengthening of ontological maturation, enlargement of the
vertebral canal and the concomitant expansion of the thoracic nerves,
shrinkage of the gastrointestinal tract, the decent of the larynx, all of
which are directly related to the emergence of language in hominid
evolution, certainly involve mutations of both structural and regulatory
genes. The kind of genetic mutation that is implausible is the kind that is
directly and exclusively responsible for the origin or the structure of
language!
Talmy mentioned in passing names of great scholars in evolution such as
Ernst Mayr. One of Mayr's seminal contributions in evolution is to point
out how behavior kicks off or initiates a long and elaborate evolutionary
process affecting complex animals such as vertebrates and mammals. For
example, the penetration of a new ecological niche by a population of
animals, which is typically initiated by behavior, often leads to dramatic
evolutionary development. (For example, cetaceans and the Darwinian finches
of the Galapagoes.) The main reason is that a new ecological niche
unleashes a new set of selectional forces. The evolutionary development
initiated by an exceptional or out-of-norm behavior such as the pentration
of a new ecological niche may involve genetic mutations favored by the new
ecological conditions. But the point is that it is often behavior, not
mutation that initiates the evolutionary development. The obvious advantage
of behaviorally initiated, as opposed to genetically-initiated,
evolutionary development is that behavior can be transmitted among social
cohorts and passed on from one generation to another. For genetic mutation
to work in the same way, it would have to occur simultaneously in at least
one male and one female of a population. Even then, the genetic mutation,
in spite of its adaptive value, may not endure because of genetic drift. If
the genetic mutation entails behavioral change among highly social animals
such as hominids, the very survival of the beneficiaries of the mutation
will be severely jeopardized. Imagine what could happen to an animal in a
group of highly social mammals if it perchance acquired a different mode of
communication through mutation!  I have postulated that the emergence of
the first symbolic signal is a behavioral innovation. That behavioral
innovation is what started the ball rolling in hominid evolution that
ultimately led to the crystallization of language, which, according to a
variety of evidence, occurred several tens of thousands of years after the
emergence of anatomically modern humans in Africa, probably during 80,000 -
60,000 before present.

I apologize for the length of this response to Talmy's repartee. As it
stands, the response contains many statements that are oversimplifications.
There are probably also many mistakes as I pecked them out at my computer
with two fingers in great haste. Talmy can rightfully snicker at my
admission of being in haste since I mentioned Givonian speed at the
beginning - "Every criticism is a piece of autobiography' (Oscar Wilde)! In
light of my heavy work load during the current California budget crisis, I
will desist from sending further replies. I will refer any reader who is
interested in my discussion to two of my recent papers both of which will
appear in print any day:

"On the evolutionary origin of language"  In Mirror neurons and the
evolution of brain and language , edited by M Stamenov and V. Gallese,
Amsterdam: John Benjamins. (Co-author Jean-Marie Hombert).

"Missing links, issues and hypotheses in the evolutionary origin of
language"  In The evolution of language and pre-language, edited by T.
Givon , Amsterdam: John Benjamins.

I might mention an in-progress article that is relevant:
"The evolutionary of basic emotions and their manifestations in
communicative behavior: Human language vs. animal communication" In
Dialogic expressions of basic emotions, edited by Edda Weigand, Amsterdam:
John Benjamins.

When I finish writing the in-progress article, I will be happy to send it
to those who request it.

Charles











>Dear FUNKnetters (and Charles Li),
>
>I have held off from responding to the various notes on the subject of
>language, culture, biology and evolution, hoping against all hope to hear
>from others what I thought needed to be said. But the more I hear, the
>more I am inclined to tear my hairs--the few still remaining--in utter
>despair. So, in the interest of  refocusing the discussion more explicitly
>towards the interface between biology, I would like to interject the
>following:
>
>The anthropocentric impulse to draw an absolute boundary between
>biological evolution ('the animal') and cultural evolution ('history',
>'the human') goes back to antiquity, and has been the subject of
>repeated  (and repetitious) bitter skirmishes ever since. Both Plato and
>Aristotle posited this absolute boundary, the latter in terms of the
>exclusively-human spirit of rationality (as distinct from the metabolic
>and perceptual spirits, which were conceded to all animate beings).
>
>The bitter Darwin-Wallace debate about evolution was about whether to draw
>that self-same rigid line between the evolution of the body and that of
>the mind/brain/soul. Wallace wanted to draw the line right where Plato,
>Aristotle, the Church, Descartes, and all traditional humanist had drawn
>it. Darwin insisted on a unitary system--what's good for the body is good
>for the soul. Chomsky's and Bickerton's insistence that the 'rise' of
>language is not governed by the same adaptive constraints that govern the
>evolution of all other biological sub-systems (including 'mere cognition')
>follows in the very same dualist tradition. And likewise, the idea that
>somehow cullture and history are wholly liberated from adaptive
>('functional') constrants, that they are--to paraphrase Charles Li (and,
>for that matter, Bill Labov)--matters of only society and culture, falls
>squarely on the Plato-Aristotle-Descartes-Wallace-Chomsky side of the debate.
>
>The funny thing is, in biology itself there have been strong recent trends
>to obliterate this rigid --artificial, ideological--boundary between
>mind/language/culture/behavior and 'plain' structural biology. Thus, a
>great evolutionary-biology theorist, Ernst Mayr, as noted that (loosely
>paraphrased) 'soft' adaptive behavior is the pacemaker, of 'hard' genetic
>evolution. And the work of great figures in etology (von Frisch, Marler,
>J.T. Bonner, E.O. Wilson, D. Griffin, F. deWaal, to mention only a few)
>has been predicated on the interpermeability of biology and culture.
>
>It therefore boggles my (admittedly simple) mind to see self-declared
>functionalists, who believe that language is adapted to its main tasks of
>representation and communication of information, come up with the idea
>that the main process that shapes both phonology and grammar--diachronic
>development--is somehow only 'culturally' and 'socially' constrained (and
>pray, what does it mean exactly? Total lack of constraints?) This
>certainly flies in the face of all detailed, fine-grained studies of the
>process of diachronic change in both phonology and grammar.
>
>"But", the defenders of the strict boundary retort, "what about
>cross-language typological variation and the great freedom from adaptive
>constraints that it entails?" In response, I would like to remind
>everybody of a few things about variation and how it comes about:
>
>First, typological variation is highly constrained. Relatively few 'types'
>of any linguistic structure are actually attested in languages--as
>compared with the plethora of mathematically (or biologically) possible
>structures. The fact that more than one type of structure can perform the
>same communicative funtion is not unique to language and culture. It is a
>cardinal fact in biology, and of the main process of functional extension
>(homoplasy), by which a structure takes over the coding of similar but
>not-wholely-identical functions (and may eventually diverge beyond
>recognition). Evolutionary biology, while adaptively guided (via
>selection) is hardly 100% deterministic, or we would have had only one
>extant species.
>
>Second, the diachronic pathways that give rise to variable synchronic
>structure are just as constrained and largely uni-directional. In both,
>they closely resemble biological evolution. And the most common
>explanation to such constrained evolution are adaptive.
>
>Third, in language diachrony as in biological evolution, the universal
>constrains that govern synchronic structure--the eplanatory principles,
>the theory--operate largely through the developmental process. They are to
>be found in the mechanisms of emergence (be they ontogenetic, phylogenetic
>of behavioral-historial, not in the catalogue of extant synchronic
>('typological') structures.
>
>Fourth, variation is the very soul of both synchronic biology and
>evolution, where species are defined by their variability curves (of both
>'hard' genotype and 'soft', 'behavioral' phenotype); where today's
>intra-species population variation is simply the current inventory of
>tomorrow's potential evolutionary changes--and thus cross-species
>variation. This is exactly what one finds in the relation between
>synchronic variation and diachronic change, an observation often
>attributed to Bill Labove (tho I cannot find an exact citation...).
>
>Lastly, variation is not only a methodological phenomenon (cf. Labov
>again); it is a highly theoretical entity in both biology and language,
>and has highly adaptive ('funtional') motivations. The reservoir of
>synchronic variants within a population is guarantor of survival, just in
>case the context should require alternative adaptive solutions.
>
>True enough, diachrony ('history') in both language and culture is 'soft'
>and presumably reversible. But so is animal behavior, which is never 100%
>constrained by genetics (cf. Mayr's "open program"). Still, this neither
>exempts language and cultrure from adaptive constraints, nor sets them
>apart from 'plain' biological evolution.
>
>It is perhaps time that we quit erecting and re-erecting those artificial
>barriers--human vs. animal, body vs. soul/mind, biology vs. culture,
>language vs. cognition--and instead re-group to take care of the urgent
>task at hand: Investigating and explaining language as and integral part
>of this incredibly complex miracle of striving, behaving, adaptive,
>evolving life. There is no shame but only strength in being part of the
>grand coalition of 'plain' biology.
>
>Have a happy holliday season,   TG
>
>=================================
>
>Charles Li wrote:
>>  Using the expression, "the evolution of language", to refer to the
>> evolution of hominid communicative behavior prior to the crystallization
>> of language implies that language was the communicative tool of all
>> hominids. Yet no one would assume that Orrorin tugenensis,
>> Kenyanthropus, the various species of Ardipithecus, Australopithecus and
>> Paranthropus had language. Indeed, most of the species in the genus of
>> Homo probably did not have language if 'language' designates the casual,
>> spoken language of anatomically modern humans. The investigation of the
>> origin of language is an enterprise concerned with the evolution of the
>> communicative behavior of our hominid ancestors, NOT the evolution of
>> language. Chronologically the study of the evolution of language begins
>> from the time when language crystallized, whereas the study of the
>> origin of language ends at the crystallization of language. This
>> distinction does not belittle the significance of the research probing
>> into older and older layers of human language. Nor does it dismiss the
>> importance of the proto-human language if and when its features can be
>> inferred. One of the most important reasons for making this distinction
>> is the fundamental difference between the ways language changes and the
>> ways hominid communicative behavior changes. The latter, like animal
>> communicative behavior, is subject to the constraint of Darwinian
>> evolution. The evolution of our hominid ancestors' communicative
>> behavior involves natural selection and genetic mutation. A change of
>> their communicative behavior in the direction toward the emergence of
>> language was adaptive in the sense that it enhanced their life
>> expectancy and reproductive success. Those hominids who made the change
>> achieved a higher level of fitness than those hominids who failed to
>> make the change. A change moving the hominids' communicative behavior
>> one step closer to human language would imply greater communicative
>> efficiency. Greater communicative efficiency would, in turn, entail
>> greater ease with which valuable knowledge and experience could be
>> passed from one individual to another and from one generation to
>> another.  Rapid and efficient transmission of knowledge conferred an
>> immense competitive advantage to the hominids for securing resources and
>> possibly vanquishing others, including other species of hominids whose
>> communicative behavior was less developed in the direction toward
>> language crystallization.
>>
>>Given that hominids within the genus of Homo and possibly some gracile
>>species of the Australopithecine are generalists who did not specialize
>>in any specific ecological niche, the competitive advantage conferred by
>>a more effective communicative behavior may explain why there is only one
>>surviving species within the taxonomic family of hominids. When two
>>hominid species happened to co-exist as generalists and the communicative
>>behavior of one species was more effective than that of the other, there
>>would be a good possibility that the communicatively more advanced
>>species would eliminate the other through competition, especially when
>>natural resources dwindled as they did periodically in a dramatic fashion
>>during the past three million years because of global temperature
>>fluctuations.
>>
>>The evolution of language, i.e. language change after its
>>crystallization, is by an large driven by social and cultural factors. It
>>has nothing to do with genetic mutation, natural selection, life
>>expectancy or reproductive success. Confusing the evolution of language
>>with the origin of language may  result in attributing features of
>>language to the communicative behavior of early hominids before the
>>emergence of language.
>>
>>For details, see attached paper.
>>
>>Charles Li
>>
>>
>>
>>At 09:36 AM 12/2/2002 +0200, Tahir Wood wrote:
>>
>>
>>> >>> Bill Croft <w.croft at MAN.AC.UK> 11/29/02 06:56PM >>>
>>>       One must distinguish between the evolution of language
>>>and the evolution of languages. The former is the evolution
>>>of human cognition and social behavior that permitted the
>>>rise of modern human language. This is of course an
>>>instance of biological evolution, of human beings. The
>>>latter is the process by which linguistic elements change
>>>over time. This is an evolutionary process, that is it
>>>involves change by replication; but it is not the same
>>>evolutionary process as biological evolution.
>>>
>>>I prefer to refer to the former as evolution and the latter as history.
>>>Less confusing, and less likely to give the impression that the social
>>>aspect of language change is absent in favour of some kind of biological
>>>determinism: e.g. Africans have 'less developed languages' because they
>>>are racially/biologically 'less developed'.
>>>Tahir
>>
>>___________________________________________________
>>Charles Li
>>Professor of Linguistics, Dean of Graduate Division
>>University of California, Santa Barbara
>>Santa Barbara, CA 93106
>>Tel: 805-893-2013               Fax: 805-893-8259
>
>___________________________________________________
>Charles Li
>Professor of Linguistics, Dean of Graduate Division
>University of California, Santa Barbara
>Santa Barbara, CA 93106
>Tel: 805-893-2013               Fax: 805-893-8259
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/funknet/attachments/20021204/78c1fd6b/attachment.htm>


More information about the Funknet mailing list